Intersection and convex combination in multi-source spectral planted cluster detection

Planted cluster detection is an important form of signal detection when the data are in the form of a graph. When there are multiple graphs representing multiple connection types, the method of aggregation can have significant impact on the results of a detection algorithm. This paper addresses the tradeoff between two possible aggregation methods: convex combination and intersection. For a spectral detection method, convex combination dominates when the cluster is relatively sparse in at least one graph, while the intersection method dominates in cases where it is dense across graphs. Experimental results confirm the theory. We consider the context of adversarial cluster placement, and determine how an adversary would distribute connections among the graphs to best avoid detection.

[1]  Dane Taylor,et al.  Enhanced detectability of community structure in multilayer networks through layer aggregation , 2015, Physical review letters.

[2]  Raj Rao Nadakuditi,et al.  Improved hidden clique detection by optimal linear fusion of multiple adjacency matrices , 2015, 2015 49th Asilomar Conference on Signals, Systems and Computers.

[3]  Jari Saramäki,et al.  Effects of time window size and placement on the structure of an aggregated communication network , 2012, EPJ Data Science.

[4]  Santo Fortunato,et al.  Community detection in graphs , 2009, ArXiv.

[5]  Steven Thomas Smith,et al.  Residuals-based subgraph detection with cue vertices , 2015, 2015 49th Asilomar Conference on Signals, Systems and Computers.

[6]  Jari Saramäki,et al.  Multiscale analysis of spreading in a large communication network , 2011, ArXiv.

[7]  Raj Rao Nadakuditi,et al.  On hard limits of eigen-analysis based planted clique detection , 2012, 2012 IEEE Statistical Signal Processing Workshop (SSP).

[8]  Patrick J. Wolfe,et al.  A Spectral Framework for Anomalous Subgraph Detection , 2014, IEEE Transactions on Signal Processing.

[9]  Alex Pentland,et al.  Reality mining: sensing complex social systems , 2006, Personal and Ubiquitous Computing.

[10]  References , 1971 .

[11]  Rajmonda Sulo Caceres,et al.  Temporal Scale of Processes in Dynamic Networks , 2011, 2011 IEEE 11th International Conference on Data Mining Workshops.

[12]  Raj Rao Nadakuditi,et al.  The eigenvalues and eigenvectors of finite, low rank perturbations of large random matrices , 2009, 0910.2120.

[13]  Frank McSherry,et al.  Spectral partitioning of random graphs , 2001, Proceedings 2001 IEEE International Conference on Cluster Computing.

[14]  Rajmonda S. Caceres,et al.  Handling Oversampling in Dynamic Networks Using Link Prediction , 2015, ECML/PKDD.

[15]  M E J Newman,et al.  Finding and evaluating community structure in networks. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[16]  E. Arias-Castro,et al.  Community Detection in Random Networks , 2013, 1302.7099.